Innovations
Neuromorphic Intelligent Visual System
■Technology Introduction
- Low-Power Processing-in-sensor CMOS Image Sensor
- Implementation of Low-Power and Low-Latency Deep Learning Chip based on Neuromorphic Intelligence.
- Development of Neuromorphic Chip based on The Fruit Fly Visual and Spatial Sensory Systems
- Hardware-Software Co-Design for Neuromorphic AI Chips.
■Scientific Innovation
- The computational CMOS image sensor (C2IS) with array-parallel computing can perform always-on feature extraction.
- Multi-precision computing in SRAM increases the DNN accuracy and performance.
- Based on real NN to design a low power object tracking neuromorphic chip with learning rule.
- FlexNet resolves the issue that incurs > 30% extra accuracy loss when changing the bit width of AI chips at run time.
■This technology won the Futuretech Breakthrough Award in 2019 Future Tech Expo.
■Institutes
National Tsing Hua University
noty102@gmail.com
|
|
本網站使用您的Cookie於優化網站及您的瀏覽經驗。繼續瀏覽網站即表示您同意本公司隱私權政策,您可至隱私權政策了解詳細資訊。